On Thu, Jan 12, 2017 at 9:15 AM, Zheng Bao fishbaoz@hotmail.com wrote:
Our VPX design uses separate reference clock source, which is Si52111-B5 (No spread), instead of common ref clock from CPU.
Now The system is unstable. Reading PCIE configuration space is unstable too. (If we add some fly wire to make it work with common ref clock, the system becomes stable.)
Can you send lspci -xxxx output from this stable common-ref-clock config?
(abstracted from PCIe spec: 12 Slot Clock Configuration – This bit indicates that the component uses the same physical reference clock that the platform provides on the connector. If the device uses an independent clock irrespective of the presence of a reference clock on the connector, this bit must be clear. For a multi-Function device, each Function must report the same value for this bit.)
Based on my understanding, the BIOS need to read bit "Slot Clock Configurationclear" to see if
separate ref clock is used. BIOS then write bit "Common Clock Configuration".
On our board, the bit "Slot Clock Configuration" is always 1, which I assume should be 0.
My question is, how the hardware affect the bit "Slot Clock Configuration"? How do we need to design our board to make the bit "Slot Clock Configuration" be 0?
Speculating below that some AMD SoC is involved in your design.
Making that bit appear as 0 is silicon specific, for You it should be possible to ask other AMD engineers directly the conditions when PCIe root ports will have this bit set as 0.
NDA'd docs from AMD that are within my reach do not have clear specs on PCIe refclock AC characteristics when running in internal clocking mode. Since PCIe separate refclock architecture has more strict requirements for clock jitter, AMD SoC's might fail there unless you are bootstrapping them to external reference clock mode.
Regards, Kyösti Mälkki
-
Kyösti Mälkki